neo4j-agent-memory-skill

Authoritative reference for the neo4j-agent-memory Python package — a graph-native memory system for AI agents built on Neo4j — and for the hosted service (NAMS) at memory.neo4jlabs.com. Use this skill whenever the user mentions neo4j-agent-memory, agent memory with Neo4j, context graphs, the POLE+O model, MemoryClient/MemorySettings, the memory MCP server, or any of the framework integrations (LangChain, PydanticAI, CrewAI, AWS Strands, Google ADK, Microsoft Agent Framework, OpenAI Agents, LlamaIndex). Also use when the user mentions the hosted service at memory.neo4jlabs.com, NAMS, the Neo4j Agent Memory Service, the `nams_` API key prefix, or the hosted MCP endpoint. Also use when writing documentation, blog posts, tutorials, PRDs, or code samples for the project, when comparing agent memory approaches, or when positioning graph-native memory against vector-only approaches — even if the user doesn't explicitly name the package.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A thorough, highly actionable reference skill with excellent executable examples and useful verify-before-publish checklists. Its main weaknesses are conciseness (overlapping checklists and corrections sections) and progressive disclosure (everything is inlined into one large file with no bundle references).

Suggestions

Consolidate the 'Quick Authoritative-Facts Checklist', 'Common Corrections to Watch For', and final 'Checklist' into a single de-duplicated checklist to remove redundancy and tighten the token budget.

Move the bulkier reference material (full extras list, NAMS surface/auth/rate-limit details, positioning language and taglines) into files under references/ and link to them one level deep, keeping SKILL.md as a lean overview.

Fill the OpenAI Agents import placeholder ('from ... import ...') with the concrete symbol so every framework integration row is copy-paste ready.

Dimension	Reasoning	Score
Conciseness	Mostly project-specific reference material that earns its place, but there is noticeable redundancy: the 'Quick Authoritative-Facts Checklist', the 'Common Corrections to Watch For' list, and the final 'Checklist' overlap heavily and could be consolidated.	3 / 5
Actionability	Copy-paste-ready code and commands throughout — the async MemoryClient quickstart, uvx MCP one-liners, Claude Code/Desktop registration configs, install/extras commands, and framework import paths cover the common cases fully.	5 / 5
Workflow Clarity	Clear structure with verify-before-publish callouts and two checklists acting as validation checkpoints, plus a 'Common Corrections to Watch For' review loop; minor gap is the absence of a single tight step sequence for the authoring workflow.	4 / 5
Progressive Disclosure	Well-sectioned with headers and tables, but the skill is a large (~22KB) monolithic SKILL.md with no bundle files in references/, scripts/, or assets/; content such as the full extras list, NAMS details, and positioning language is inlined rather than split into one-level-deep references.	3 / 5
	Total	15 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An exceptionally complete and specific description that clearly states what the skill is, gives explicit 'Use this skill whenever…' trigger guidance, and packs in package-specific synonyms and trigger terms. No fluff or vague language.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities — 'writing documentation, blog posts, tutorials, PRDs, or code samples', 'comparing agent memory approaches', 'positioning graph-native memory against vector-only approaches' — with comprehensive coverage of the package's surface.	5 / 5
Completeness	Explicitly answers both what ('Authoritative reference for the neo4j-agent-memory Python package — a graph-native memory system…') and when ('Use this skill whenever the user mentions…', with several concrete trigger scenarios, even when the package is not named).	5 / 5
Trigger Term Quality	Comprehensive natural-term coverage including synonyms ('neo4j-agent-memory', 'NAMS', 'Neo4j Agent Memory Service', 'hosted service at memory.neo4jlabs.com', 'nams_ API key prefix', 'context graphs', 'POLE+O model') plus all framework names.	5 / 5
Distinctiveness Conflict Risk	Scoped to a single named package and its hosted service with package-specific triggers (POLE+O, nams_ prefix, memory.neo4jlabs.com), giving it a clear niche with minimal conflict risk.	5 / 5
	Total	20 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: neo4j-contrib/neo4j-skills
Path: neo4j-agent-memory-skill/SKILL.md
Commit: 02ad354

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.